If you would carefully look at "Best headphones" example, the reddit result card shows '2 days ago', whereas the link it points to is 7 months old.
Similarly the Sennheiser headphones results card shows $379 on Amazon, when its actually $400 today.
No offenses to how well it works, but if I had to be sold to get a subscription, I would rather like to see a real-time example. A canned example FWIW could be a completely scripted search result.
I am personally not bothered by small errors here and there, it is important to get the big picture right - alignment of incentives inside the search experience. Overall, I believe we also have superior results to Google, please try it for a few days and share your thoughts.
Kagi is in some ways broken and flawed and it is what makes it feel more humane to use.
If there are things that particulary bother you feel free to share them on kagifeedback.org. We are not ignorant of these, just limited by our current resources.
To be clear, I pay $20/mo for ProtonMail Visionary so I'm not averse to paying! I just can't see the value at $30/mo when the other family members don't use it nearly as much as I do.
Maybe this is an untenable problem due to your costs and you need those low use paying users. I'm not sure.
To clarify: by intermixed I mean our tuning was different so it became somewhat interesting, as we all wanted to tune the settings differently. It's not a huge thing, but between that and feeling bad for having three people using one sub...
> We are a ten people bootstrapped team.
We are a 2-engineer startup. We give business analytics/market trends on mined webdata (through our B2B dashboards) with a 48 hour live data guarantee. Every time, user queries crunch through about 9 million data points for every day over last 36 months - in real time. Sorry, but it isn't a very good reason to have stale results.
The fact that it lists ballpark prices rather than exact prices is really not that important in the grand scheme of things.
Also kind of a dick move to compare to your startup to Kagi to justify it sucking. Analytics is very different to search. Your products sound completely different.
Their example is the headphone search. That Reddit post is a few months old, and has _one_ comment on it. The thread asks specifically about closed-back headphones under $150.
Something like this would be much better as a result: https://www.reddit.com/r/headphones/comments/nq0pt/headphone...
Remember we are comparing this to Google where the entire first page for this query is either ads or results full of ads and affiliate links.
Please don't compare yourself to Google. Most people on HN are well aware that Google search has become a terrible baseline. DDG, appending Reddit to query etc., are workarounds discussed a lot of times on the forum.
I have no beef against Kagi. I wish you guys succeed, but please have a better product argument. At this point, I actually find equally good suggestion on 'headphones' from both DDG & Ecosia. If Kagi needs to win over, it should be doing better vis-a-vis the upcoming search engines, not Google which has been SEO gamed over & over.
This happens on Google too, all of the time. I think this is a reddit issue, with how they have their meta configured...
